First I need to load up the packages I’ll need
library(sf)
## Linking to GEOS 3.4.2, GDAL 2.1.2, proj.4 4.9.1
library(ggplot2) #development version!
## devtools::install_github("tidyverse/ggplot2")
library(tidyverse)
## Loading tidyverse: tibble
## Loading tidyverse: tidyr
## Loading tidyverse: readr
## Loading tidyverse: purrr
## Loading tidyverse: dplyr
## Conflicts with tidy packages ----------------------------------------------
## filter(): dplyr, stats
## lag(): dplyr, stats
library(readr)
## Not sure about this bit
#library("tidyverse",lib.loc="/Library/Frameworks/R.framework/Versions/3.4/Resources/library")
library(cowplot)
##
## Attaching package: 'cowplot'
## The following object is masked from 'package:ggplot2':
##
## ggsave
library(sp)
library(gridExtra)
##
## Attaching package: 'gridExtra'
## The following object is masked from 'package:dplyr':
##
## combine
library(dplyr)
library(ggrepel)
library(plyr)
## -------------------------------------------------------------------------
## You have loaded plyr after dplyr - this is likely to cause problems.
## If you need functions from both plyr and dplyr, please load plyr first, then dplyr:
## library(plyr); library(dplyr)
## -------------------------------------------------------------------------
##
## Attaching package: 'plyr'
## The following objects are masked from 'package:dplyr':
##
## arrange, count, desc, failwith, id, mutate, rename, summarise,
## summarize
## The following object is masked from 'package:purrr':
##
## compact
Now I import my data. I filter for the Arran postcodes, (since Arran all begins ‘KA27’).
## Finding the Arran coordinates
arrancoordinates <- read.csv("alldata/ukpostcodes.csv") %>%
filter(substr(postcode,1,4)=="KA27")
pcs <- read_sf("alldata/Scotland_pcs_2011")
arransubsect <- filter(pcs,substr(label,1,4)=="KA27")
Now I load the SIMD data, containing the geometries (shapefiles) and SIMD data (percentiles, etc)
#Import SIMD data from http://www.gov.scot/Topics/Statistics/SIMD
#https://data.gov.uk/dataset/scottish-index-of-multiple-deprivation-simd-2012
#https://data.gov.uk/dataset/scottish-index-of-multiple-deprivation-simd-2012/resource/d6fa8924-83da-4e80-a560-4ef0477f230b
DZBoundaries2016 <- read_sf("./alldata/SG_SIMD_2016")
DZBoundaries2012 <- read_sf("./alldata/SG_SIMD_2012")
DZBoundaries2009 <- read_sf("./alldata/SG_SIMD_2009")
DZBoundaries2006 <- read_sf("./alldata/SG_SIMD_2006")
DZBoundaries2004 <- read_sf("./alldata/SG_SIMD_2004")
I have to choose the right columns manually in order to select the Arran data.
#Selecting Arran data from Scotland (2016)
#Find postcode look-up from below file for KA27 postcodes. Find unique DZ. Find row positions.
#SIMD2016 <-read.csv("./alldata/00505244.csv")
#Selecting ArranDZ2016
Arrandz2016 <- c(4672,4666,4669,4671,4667,4668,4670)
arran2016 <- DZBoundaries2016[Arrandz2016,]
#Reorder arran 2016
reorderedvector<- c("S01011174", "S01011171", "S01011177", "S01011176", "S01011175", "S01011173", "S01011172" )
arran2016 <- arran2016 %>%
slice(match(reorderedvector, DataZone))
#Find postcode look-up, KA27 postcodes. Find unique DZ. Find row positions.
#Selecting ArranDZ2012
Arrandz2012 <- c(4409,4372,4353,4352,4351,4350,4349)
#2012
arran2012 <- DZBoundaries2012[Arrandz2012,]
#2009
arran2009 <- DZBoundaries2009[Arrandz2012,]
#2006
arran2006 <- DZBoundaries2006[Arrandz2012,]
#2004
arran2004 <- DZBoundaries2004[Arrandz2012,]
Now I want to plot all the data, first I combine it all into one table. First I subselect the data I want from the appropriate columns.
arran20162 <- arran2016 %>%
select(DataZone, geometry, Percentile) %>%
mutate(year="2016")
arran20122 <- arran2012 %>%
select(DataZone, geometry, Percentile) %>%
mutate(year="2012")
arran20092 <- arran2009 %>%
select(DataZone, geometry, Percentile) %>%
mutate(year="2009")
arran20062 <- arran2006 %>%
select(DataZone, geometry, Percentile) %>%
mutate(year="2006")
arran20042 <- arran2004 %>%
select(DataZone, geometry, Percentile) %>%
mutate(year="2004")
#Now I add it together
arransimd <- rbind(arran20162,arran20122,arran20092,arran20062,arran20042)
The reason I’ve downloaded all the datazones shapefiles individually is because they change between 2016 and 2012. See the small differences.
arran1612 <- rbind(arran20162,arran20122)
arran1612 %>%
ggplot() +
geom_sf(aes(fill = DataZone)) +
facet_wrap('year') +
theme_grey() +
theme(legend.position="none") +
theme(axis.text.x=element_text(angle=45, hjust = 1))
There we are. The SIMD health percentiles of Arran zones throughout SIMD history. And I’ve learned a little bit about graphics in R.
If I wanted to I could show the zones individually.. First I find the unique zones. (There are 14. 7 Zones 2016, 7 Zones pre-2016)
datazones <- unique(arransimd$DataZone)
I’ll have to find out a simpler way to do this but.. In order to turn the names into arguments I’ve made a function that filters the data into an individual name. #Pre-2016 Individual Zones
function0.5 <- function(argument)
{
filter(arransimd, DataZone==argument)
}
So by reading ‘datazones’ I’ve made a list of the output
#all datazones
datazonelist <- lapply(datazones, function0.5)
#Pre-2016 lists
pre2016list2 <- list("S01004409", "S01004372", "S01004353", "S01004352", "S01004351", "S01004350", "S01004349")
#Create a new way of making character list
pre2016list <- lapply(pre2016list2, function0.5)
post2016list2 <- list("S01011174", "S01011171", "S01011177", "S01011176", "S01011175", "S01011173", "S01011172")
post2016list <- lapply(post2016list2, function0.5)
Now I want to overlay the postcodes for a particular shapefile, in this case by Datazone. To do this I’ve converted both the Arran coordinates and Arran (2016) shapefiles into Spatial Points/Polygons, converted them into a common CRS, and then compared them by using over().
simple.sf <- st_as_sf(arrancoordinates, coords=c('longitude','latitude'))
st_crs(simple.sf) <- 4326
exampleshapes <- sf:::as_Spatial(arran2016$geometry)
examplepoints <- sf:::as_Spatial(simple.sf$geom)
examplepoints <- spTransform(examplepoints, CRS("+proj=longlat +datum=WGS84"))
exampleshapes <- spTransform(exampleshapes, CRS("+proj=longlat +datum=WGS84"))
namingdzpostcode <- over(exampleshapes, examplepoints, returnList = TRUE)
I can then take a member reference from the orginal postcode list, which gives me a selection of the rows in that DZ. For simplicity I’ve written this as a new function.
Unfortunately, I haven’t worked out how to coordinate the new ID with the original DZ names yet, so I have to select by using the appropriate ID for each DZ.
pre2016listID <- list(1,2,3,4,5,6,7)
post2016listID <- list(1,2,3,4,5,6,7)
listID <- list(1,2,3,4,5,6,7)
function6 <- function(argument)
{
arrancoordinates[namingdzpostcode[[argument]],]
}
function100 <- function(argument)
{
argument <- arrancoordinates[namingdzpostcode[[argument]],] %>% mutate(DataZone=argument)
}
function100(1)
newarrancoordinates <- lapply(1:7,function100)
newarrancoordinates <- rbind(newarrancoordinates[[1]], newarrancoordinates[[2]], newarrancoordinates[[3]], newarrancoordinates[[4]], newarrancoordinates[[5]], newarrancoordinates[[6]], newarrancoordinates[[7]])
newarrancoordinates$listID <- revalue(as.character(newarrancoordinates$DataZone),
c('1'="S01004409/S01011174", '2'="S01004372/S01011171", '3'="S01004353/S01011177", '4'="S01004352/S01011176", '5'="S01004351/S01011175", '6'="S01004350/S01011173", '7'="S01004349/S01011172"))
I’ve also made another function to plot the DZ on it’s own with coordinates.
function5 <- function(argument, argument2)
{
argument %>%
ggplot() +
geom_sf() +
theme_grey() +
geom_point(data=function6(argument2), mapping = aes(x = longitude, y = latitude), size=1) +
theme(axis.title.x = element_blank(),
axis.title.y = element_blank()) +
coord_sf(crs= 4326, datum = sf::st_crs(4326))
}
function5.5 <- function(argument, argument2)
{
argument %>%
ggplot() +
geom_sf() +
theme_grey() +
geom_point(data=function6(argument2), mapping = aes(x = longitude, y = latitude), size=1) +
geom_text_repel(data = function6(2),
aes(label = function6(2)$postcode, x = longitude, y = latitude), size=2) +
theme(axis.title.x = element_blank(),
axis.title.y = element_blank()) +
coord_sf(crs= 4326, datum = sf::st_crs(4326))
}
function1.5.5 <- function(argument)
{
argument %>%
mutate(
lon = map_dbl(geometry, ~st_centroid(.x)[[1]]),
lat = map_dbl(geometry, ~st_centroid(.x)[[2]])
) %>%
ggplot() +
geom_sf(aes(fill = Percentile)) +
facet_wrap('year') +
theme_grey() +
geom_text(aes(label = Percentile, x = lon, y = lat), size = 2, colour = "white") +
theme(axis.text.x=element_blank(),
axis.ticks.x=element_blank(),
axis.title.x = element_blank(),
axis.title.y = element_blank()) +
theme(legend.position="bottom")
}
function2.5.1 <- function(argument)
{
arransubsect %>%
ggplot() +
geom_sf() +
theme_grey() +
theme(axis.text.x=element_text(angle=45, hjust = 1)) +
theme(legend.position="bottom") +
geom_sf(data= argument, aes(fill = DataZone))
}
function7.5.1 <- function(argument, argument2)
{
a <- function1.5.5(argument)
b <- function2.5.1(argument)
c <- function5(argument, argument2)
grid.arrange(a, b, c, nrow = 1)
}
function8.pre <- function(argument)
{
function7.5.1(pre2016list[[argument]],listID[[argument]])
}
lapply(1:7, function8.pre)
## [[1]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[2]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[3]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[4]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[5]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[6]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[7]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
function8.post <- function(argument)
{
function7.5.1(post2016list[[argument]],listID[[argument]])
}
lapply(1:7, function8.post)
## [[1]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[2]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[3]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[4]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[5]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[6]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
##
## [[7]]
## TableGrob (1 x 3) "arrange": 3 grobs
## z cells name grob
## 1 1 (1-1,1-1) arrange gtable[layout]
## 2 2 (1-1,2-2) arrange gtable[layout]
## 3 3 (1-1,3-3) arrange gtable[layout]
overlay info onto leaflet then with labels.